Abstract
Many problems can be formulated as recovering a low-rank tensor. Although an increasingly common task, tensor recovery remains a challenging problem because of the delicacy associated with the decomposition of higher-order tensors. To overcome these difficulties, existing approaches often proceed by unfolding tensors into matrices and then apply techniques for matrix completion. We show here that such matricization fails to exploit the tensor structure and may lead to suboptimal procedure. More specifically, we investigate a convex optimization approach to tensor completion by directly minimizing a tensor nuclear norm and prove that this leads to an improved sample size requirement. To establish our results, we develop a series of algebraic and probabilistic techniques such as characterization of subdifferential for tensor nuclear norm and concentration inequalities for tensor martingales, which may be of independent interests and could be useful in other tensor-related problems.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
E.J. Candès and B. Recht, Exact matrix completion via convex optimization, Foundations of Computational Mathematics, 9, 717–772 (2008).
E.J. Candès and T. Tao, The power of convex relaxation: Near-optimal matrix completion, IEEE Transactions on Information Theory, 56(5), 2053–2080 (2009).
S. Cohen and M. Collins, Tensor decomposition for fast parsing with latent-variable pcfgs, in Proceedings of NIPS 2012 (2012).
S. Gandy, B. Recht and I. Yamada, Tensor completion and low-n-rank tensor recovery via convex optimization, Inverse Problems, 27(2), 025010 (2011).
E. Giné and J. Zinn, Some limit theorems for empirical processes, The Annals of Probability, 12(4), 929–989 (1984).
D. Gross, Recovering low-rank matrices from few coefficients in any basis, IEEE Transaction on Information Theory, 57, 1548–1566 (2011).
C. Hillar and L.H. Lim, Most tensor problems are NP-hard, Journal of the ACM, 60(6), Art. 45 (2013).
T.G. Kolda and B.W. Bader, Tensor decompositions and applications, SIAM Review, 51(3), 455–500 (2009).
J.B. Kruskal, Rank, decomposition, and uniqueness for 3-way and N-way arrays, in “Multiway data analysis”, North-Holland, Amsterdam, pp. 7–18 (1989).
N. Li and B. Li, Tensor completion for on-board compression of hyperspectral images, In Image Processing (ICIP), 2010 17th IEEE International Conference on, 517–520 (2010).
J. Liu, P. Musialski, P. Wonka and J. Ye, Tensor completion for estimating missing values in visual data, In ICCV, 2114–2121 (2009).
N. Mesgarani, M. Slaney and S. Shamma, Content-based audio classification based on multiscale spectro-temporal features, IEEE Transactions on Speech and Audio Processing, 14(3), 920–930 (2006).
C. Mu, B. Huang, J. Wright and D. Goldfarb, Square deal: lower bounds and improved relaxations for tensor recovery, arXiv: 1307.5870 (2013).
B. Recht, A simpler approach to matrix completion, Journal of Machine Learning Research, 12, 3413–3430 (2011).
O. Semerci, N. Hao, M. Kilmer and E. Miller, Tensor based formulation and nuclear norm regularizatin for multienergy computed tomography, IEEE Transactions on Image Processing, 23(4), 1678–1693 (2014).
N.D. Sidiropoulos and N. Nion, Tensor algebra and multi-dimensional harmonic retrieval in signal processing for mimo radar, IEEE Trans. on Signal Processing, 58(11), 5693–5705 (2010).
M. Signoretto, L. De Lathauwer and J. Suykens, Nuclear norms for tensors and their use for convex multilinear estimation (2010). Preprint at ftp://ftp.esat.kuleuven.be/SISTA/signoretto/Signoretto_nucTensors.
M. Signoretto, R. Van de Plas, B. De Moor and J. Suykens, Tensor versus matrix completion: A comparison with application to spectral data, IEEE SPL, 18(7), 403–406 (2011).
R. Tomioka, K. Hayashi and H. Kashima, Estimation of low-rank tensors via convex optimization, arXiv preprint arXiv:1010.0789 (2010).
R. Tomioka, T. Suzuki, K. Hayashi and H. Kashima, Statistical performance of convex tensor decomposition, Advances in Neural Information Processing Systems (NIPS), 137 (2011).
J. Tropp, User-friendly tail bounds for sums of random matrices, Foundations of Computational Mathematics, 12, 389–434 (2012).
G. A. Watson, Characterization of the subdifferential of some matrix norms, Linear Algebra Appl., 170, 33–45 (1992).
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Emmanuel Candès.
The research of Ming Yuan was supported in part by NSF Career Award DMS-1321692 and FRG Grant DMS-1265202, and NIH Grant 1U54AI117924-01. The research of Cun-Hui Zhang was supported in part by NSF Grants DMS-1129626 and DMS-1209014.
Rights and permissions
About this article
Cite this article
Yuan, M., Zhang, CH. On Tensor Completion via Nuclear Norm Minimization. Found Comput Math 16, 1031–1068 (2016). https://doi.org/10.1007/s10208-015-9269-5
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10208-015-9269-5
Keywords
- Concentration inequality
- Convex optimization
- Dual certificate
- Matrix completion
- Nuclear norm minimization
- Subdifferential
- Tensor completion
- Tensor rank